116
7 Communication
signaling and nerve transmission. However, no Newton of biology will ever come
with precise quantitative laws. Perhaps the only general theoretical principle applicable to all communication networks is the necessity of combining activating and
inhibiting signals. We can only guess whether Turing (1952) realized the general
significance of this principle in his celebrated paper, which treated a very specific
problem and ended on a sad admission that biological phenomena are very complicated (Sect. 3.4). Well, we know they are. He himself helped to activate his country’s
eventual victory by inhibiting German communications, and, in return for this service, was dismally inhibited by the homophobic establishment of the time.
Notwithstanding all the complexity being unraveled piece by piece by biologists,
theorists cannot avoid the temptation to model life, and not just on the level of
the Game of Life (Sect. 2.5). In Sect. 5.4, we talked about the dynamic tangle of
the cytoskeleton and in Sect. 5.6 about the clumsy way things crawl. Can this be
modeled by applying well established mechanical laws? Yes, certainly, and it has
been attempted not once, but the problem is that it is possible to imitate superficial
features of observed phenomena by constructing a model that has nothing to do with
reality. Attempted models of the mechanics of cytoskeletal rearrangements and cell
motion range from the most detailed, imitating the motion of every actin monomer
and every myosin motor in the network on a voracious computer, to the simplest
coarse-grained models. The former approach cannot be realistically extended to an
entire cell, and even on this level, molecular detail, including intricate chemical
signaling, cannot be accounted for.
The most sophisticated coarse-grained model is poroelastic, originating in the
theory of fluid-saturated porous soils (Biot, 1941), which applies separate mechanical equations to the elastic filament network and viscous cytosol. This model leaves
aside the permanent restructuring of the live actin–myosin network, as well as the
crowded cytosol environment, but it is still computationally difficult, and contains
unreliably measured parameters. Simplification goes further, and with a brighter illusion of success. The magic word here is “activity”. By treating the interior of a cell
as an “active medium” with whatever properties might be helpful, whether viscous
or elastic, and polarized in a suitable way, any kind of motion can be imitated. It is
rather easy to use one of the modifications of generic models to produce a moving
active spot and interpret it as a moving cell, and, whenever needed, add details to
better imitate reality. However, a visible correspondence with experiment does not
mean that the mechanism built into the model is indeed the one operating in reality.
Recall the model of multiplying droplets (Zwicker et al, 2017) in Sect. 4.4. It is
simple and elegant but I doubt that the authors seriously thought that the first selfreplicating cells might have been formed without being protected by a surfactant
sheath.
Mechanical theories become more reliable on a macroscopic level, considering
individual cells as elementary units, either incorporated in a tissue or crawling on a
substrate, but here again, the difficulty is in incorporating chemical signaling. Modeling is more reliable in biomorphic processes involving soft materials but lacking
the intricate interactions of living cells. We come to this in the next, concluding
chapter.
7 Communication
signaling and nerve transmission. However, no Newton of biology will ever come
with precise quantitative laws. Perhaps the only general theoretical principle applicable to all communication networks is the necessity of combining activating and
inhibiting signals. We can only guess whether Turing (1952) realized the general
significance of this principle in his celebrated paper, which treated a very specific
problem and ended on a sad admission that biological phenomena are very complicated (Sect. 3.4). Well, we know they are. He himself helped to activate his country’s
eventual victory by inhibiting German communications, and, in return for this service, was dismally inhibited by the homophobic establishment of the time.
Notwithstanding all the complexity being unraveled piece by piece by biologists,
theorists cannot avoid the temptation to model life, and not just on the level of
the Game of Life (Sect. 2.5). In Sect. 5.4, we talked about the dynamic tangle of
the cytoskeleton and in Sect. 5.6 about the clumsy way things crawl. Can this be
modeled by applying well established mechanical laws? Yes, certainly, and it has
been attempted not once, but the problem is that it is possible to imitate superficial
features of observed phenomena by constructing a model that has nothing to do with
reality. Attempted models of the mechanics of cytoskeletal rearrangements and cell
motion range from the most detailed, imitating the motion of every actin monomer
and every myosin motor in the network on a voracious computer, to the simplest
coarse-grained models. The former approach cannot be realistically extended to an
entire cell, and even on this level, molecular detail, including intricate chemical
signaling, cannot be accounted for.
The most sophisticated coarse-grained model is poroelastic, originating in the
theory of fluid-saturated porous soils (Biot, 1941), which applies separate mechanical equations to the elastic filament network and viscous cytosol. This model leaves
aside the permanent restructuring of the live actin–myosin network, as well as the
crowded cytosol environment, but it is still computationally difficult, and contains
unreliably measured parameters. Simplification goes further, and with a brighter illusion of success. The magic word here is “activity”. By treating the interior of a cell
as an “active medium” with whatever properties might be helpful, whether viscous
or elastic, and polarized in a suitable way, any kind of motion can be imitated. It is
rather easy to use one of the modifications of generic models to produce a moving
active spot and interpret it as a moving cell, and, whenever needed, add details to
better imitate reality. However, a visible correspondence with experiment does not
mean that the mechanism built into the model is indeed the one operating in reality.
Recall the model of multiplying droplets (Zwicker et al, 2017) in Sect. 4.4. It is
simple and elegant but I doubt that the authors seriously thought that the first selfreplicating cells might have been formed without being protected by a surfactant
sheath.
Mechanical theories become more reliable on a macroscopic level, considering
individual cells as elementary units, either incorporated in a tissue or crawling on a
substrate, but here again, the difficulty is in incorporating chemical signaling. Modeling is more reliable in biomorphic processes involving soft materials but lacking
the intricate interactions of living cells. We come to this in the next, concluding
chapter.